CN111898486B - Monitoring picture abnormality detection method, device and storage medium - Google Patents
Monitoring picture abnormality detection method, device and storage medium Download PDFInfo
- Publication number
- CN111898486B CN111898486B CN202010676575.0A CN202010676575A CN111898486B CN 111898486 B CN111898486 B CN 111898486B CN 202010676575 A CN202010676575 A CN 202010676575A CN 111898486 B CN111898486 B CN 111898486B
- Authority
- CN
- China
- Prior art keywords
- frame image
- monitoring
- current frame
- picture
- video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 208
- 238000001514 detection method Methods 0.000 title claims abstract description 92
- 230000005856 abnormality Effects 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 38
- 230000008859 change Effects 0.000 claims description 72
- 238000012806 monitoring device Methods 0.000 claims description 38
- 238000004590 computer program Methods 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000005316 response function Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
- G06T7/248—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The application provides a detection method, a device and a storage medium for monitoring picture abnormality. The method comprises the following steps: acquiring a current frame image shot by monitoring equipment and a background frame image for carrying out anomaly detection on a monitoring picture of the monitoring equipment; and carrying out anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points, and obtaining a picture anomaly detection result.
Description
Technical Field
The present application relates to the field of video image processing technologies, and in particular, to a method and apparatus for detecting an abnormality in a monitoring picture, and a storage medium.
Background
With the proposal and popularization of the concepts of safe city and smart city, the security market has shown a rapid growth of geometry in recent years.
However, along with the rapid growth of security video monitoring, the application of video monitoring is often affected by external factors, so that video monitoring cannot be effectively implemented, for example, abnormal situations such as video shielding, scene change and the like may occur in a monitoring picture of a monitoring video, which are all problems that have to be faced and have to be solved in the development process of security video monitoring.
Disclosure of Invention
One or more embodiments of the present application provide a method, an apparatus, and a storage medium for detecting an anomaly of a monitoring picture, so as to at least solve the problem of how to find out in time how to perform an anomaly phenomenon such as video occlusion, scene change, etc. on the monitoring picture.
According to an embodiment of the present application, there is provided a method for detecting an abnormality of a monitoring screen, including: acquiring a current frame image shot by monitoring equipment and a background frame image for carrying out anomaly detection on a monitoring picture of the monitoring equipment; and carrying out anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points, and obtaining a picture anomaly detection result.
In at least one exemplary embodiment, performing anomaly detection on a monitoring screen of the monitoring device according to a texture difference and/or a position distance of pairing feature points between the current frame image and the background frame image includes at least one of: determining whether a video shielding phenomenon occurs to a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image; and determining whether a scene change phenomenon occurs in a monitoring picture of the monitoring equipment according to the position distance of the pairing feature points between the current frame image and the background frame image.
In at least one exemplary embodiment, determining whether a video occlusion phenomenon occurs in a monitoring screen of the monitoring device according to the texture difference between the current frame image and the background frame image includes: extracting textures corresponding to the current frame image and textures corresponding to the background frame image according to the edge gradients of the current frame image and the background frame image respectively; and under the condition that the texture corresponding to the current frame image is smaller than the texture corresponding to the background frame image, determining that a video shielding phenomenon occurs in a monitoring picture of the monitoring equipment.
In at least one exemplary embodiment, the texture corresponding to the current frame image is smaller than the texture corresponding to the background frame image includes: the texture X c corresponding to the current frame image and the texture X b corresponding to the background frame image meet the condition X b-Xc≥θ1, wherein θ 1 is a texture difference judgment threshold value and the value is larger than 0.
In at least one exemplary embodiment, in a case where it is determined that the video occlusion phenomenon occurs in the monitoring screen of the monitoring device, the method further includes: acquiring data X' m of a frame of image shot before a video shielding phenomenon occurs on a monitoring picture of the monitoring equipment; sequentially determining a difference value eta=x ' m-X'j between data X ' j and data X ' m of the N frames of images after the video occlusion phenomenon occurs, wherein j=1, 2, … N, and N is a preset number; and filtering out a determination result of the video shielding phenomenon of a monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that a difference value eta corresponding to the picture exists in the N frames of pictures and is smaller than a preset difference value threshold.
In at least one exemplary embodiment, determining whether a scene change phenomenon occurs in a monitoring screen of the monitoring device according to the position distance of the pairing feature point between the current frame image and the background frame image includes: extracting a first characteristic point in the current frame image and a second characteristic point in the background frame image; calculating a first feature descriptor of each first feature point in a preset area corresponding to each first feature point in the current frame image, and calculating a second feature descriptor of each second feature point in a preset area corresponding to each second feature point in the background frame image; determining pairing feature points between the current frame image and the background frame image according to the first feature descriptors and the second feature descriptors; and under the condition that the distance between the pairing feature points between the current frame image and the background frame image is larger than a preset distance threshold value, determining that a scene change phenomenon occurs in a monitoring picture of the monitoring equipment.
In at least one exemplary embodiment, determining that the distance of the paired feature points between the current frame image and the background frame image is greater than a predetermined distance threshold comprises: determination ofWherein P j,c and P j,b are the positions of paired feature points between the current frame image and the background frame image, respectively, and K is the total logarithm of the paired feature points.
In at least one exemplary embodiment, in a case where it is determined that a scene change phenomenon occurs in a monitoring screen of the monitoring device, the method further includes: and acquiring a current frame image with the definition higher than a preset definition threshold, taking the acquired current frame image as a background frame image, and returning to execute the operation of determining whether the scene change phenomenon occurs on a monitoring picture of the monitoring equipment according to the position distance of the pairing characteristic points between the current frame image and the background frame image.
In at least one exemplary embodiment, in a case where it is determined that a scene change phenomenon occurs in a monitoring screen of the monitoring device, the method further includes: respectively expanding textures corresponding to the current frame image and the background frame image based on an expansion edge algorithm; determining the expansion result of the texture corresponding to the background frame imageExpansion result/>, of texture corresponding to the current frame imageDifference between/>And filtering out a determination result of the scene change phenomenon of the monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that the difference value rho is smaller than a preset expansion threshold value.
In at least one exemplary embodiment, in case that it is determined that the video occlusion phenomenon or the scene change phenomenon occurs on a monitoring screen of the monitoring device, the method further includes: determining whether the number of consecutive frame images in which the video occlusion phenomenon or the scene change phenomenon is detected to occur exceeds a predetermined frame number threshold; and filtering out the determination result of the video shielding phenomenon or the scene change phenomenon of the monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that the number of continuous frame images does not exceed the preset frame number threshold value.
According to another embodiment of the present application, there is provided a detection apparatus for monitoring screen abnormality, including: the acquisition module is used for acquiring a current frame image shot by the monitoring equipment and a background frame image used for carrying out anomaly detection on a monitoring picture of the monitoring equipment; the anomaly detection module is used for carrying out anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points, and obtaining a picture anomaly detection result.
In at least one example embodiment, the anomaly detection module includes: the video shielding detection sub-module is used for determining whether a video shielding phenomenon occurs on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image; and the scene change detection sub-module is used for determining whether a scene change phenomenon occurs in a monitoring picture of the monitoring equipment according to the position distance of the pairing characteristic points between the current frame image and the background frame image.
According to a further embodiment of the application, there is also provided a storage medium having stored therein a computer program, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.
According to a further embodiment of the application there is also provided a monitoring device or a monitoring platform or an intelligent video monitoring server comprising a memory and a processor, the memory having stored therein a computer program arranged to run the computer program to perform the steps of any of the method embodiments described above.
According to still another embodiment of the present application, there is also provided a detection system for monitoring screen abnormality, including: the one or more monitoring devices are arranged to collect monitoring video streams and transmit the monitoring video streams to the video monitoring server; the video monitoring server is connected with the one or more monitoring devices and comprises a memory and a processor, wherein the memory stores a computer program, and the processor is used for running the computer program to execute the following operations: receiving the monitoring video stream transmitted by the monitoring equipment; acquiring a current frame image in the monitoring video stream and a background frame image for carrying out anomaly detection on a monitoring picture of the monitoring equipment; and carrying out anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points, and obtaining a picture anomaly detection result.
According to one or more embodiments of the present application, since anomaly detection is performed on the monitoring picture of the monitoring device according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points, anomaly conditions of the monitoring device can be rapidly identified based on the texture difference and the position distance of the pairing feature points, so that problems of how to timely find out anomaly phenomena (such as video occlusion, scene change, etc. (e.g., artificial or unexpected noise occlusion or scene change phenomenon) of the monitoring picture can be solved, and timely and rapid detection of the anomaly of the monitoring picture can be realized.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
Fig. 1 is a flowchart of a detection method of monitoring screen abnormality according to embodiment 1 of the present application;
fig. 2 is a block diagram of a configuration of a detection apparatus for monitoring screen abnormality according to embodiment 2 of the present application;
fig. 3 is a detailed block diagram of the structure of a detection device for monitoring screen abnormality according to embodiment 2 of the present application;
Fig. 4 is a detailed flowchart of a detection method of monitoring screen abnormality according to embodiment 4 of the present application.
Detailed Description
The intelligent video monitoring technology attracts more and more attention of researchers, however, the development of the technology encounters various restrictions, and one of the restrictions is shielding target tracking. In video monitoring of a single camera, mutual occlusion between targets is a common phenomenon due to observation angles and the like, which causes serious occlusion to a real-time monitoring camera target area and even affects the use of a video monitoring system. How to use intelligent video monitoring technology to diagnose the monitoring picture abnormality, detect the shielding and scene change phenomenon of the monitoring equipment, which is artificial or unexpected noise, and has great significance for timely processing and repairing fault equipment.
The application will be described in detail hereinafter with reference to the drawings in conjunction with embodiments. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order.
Example 1
The method embodiment provided by the embodiment 1 of the application can be realized by monitoring equipment, a monitoring platform or an intelligent video monitoring server.
In this embodiment, a method for detecting an abnormality of a monitoring screen is provided, fig. 1 is a flowchart of a method for detecting an abnormality of a monitoring screen according to embodiment 1 of the present application, as shown in fig. 1, the flowchart includes the steps of:
Step S102, acquiring a current frame image shot by monitoring equipment and a background frame image used for carrying out anomaly detection on a monitoring picture of the monitoring equipment;
Step S104, carrying out anomaly detection on the monitoring picture of the monitoring equipment according to the texture difference and/or the position distance of the pairing feature points between the current frame image and the background frame image, and obtaining a picture anomaly detection result.
By the steps, the abnormal condition of the monitoring device can be rapidly identified based on the texture difference and the position distance of the paired characteristic points because the abnormal detection is carried out on the monitoring picture of the monitoring device according to the texture difference and/or the position distance of the paired characteristic points between the current frame image and the background frame image, so that the problem of how to timely find out abnormal phenomena (such as video shielding, scene change and the like (for example, shielding of artificial or unexpected noise or scene change phenomenon) of the monitoring picture is solved, and the timely and rapid detection of the abnormal condition of the monitoring picture is realized.
Alternatively, the main body of execution of the above steps may be a monitoring device, a monitoring platform, or an intelligent video monitoring server, but is not limited thereto.
In at least one example embodiment, step S104 may include at least one of:
Step S104-1, determining whether a video shielding phenomenon occurs in a monitoring picture of the monitoring device according to the texture difference between the current frame image and the background frame image, by which whether video shielding exists or not can be determined based on the texture difference of the current frame relative to the background frame, for example, if the texture difference is large (much smaller), the video shielding is indicated;
step S104-2, determining whether a scene change phenomenon occurs in a monitoring screen of the monitoring device according to the position distance between the paired feature points between the current frame image and the background frame image, by which whether a scene change occurs can be determined by the position distance between the paired feature points, for example, when the position distance between the paired feature points is large, it indicates that a deflection occurs in the monitoring screen (or monitoring field of view), which indicates that a scene change occurs.
In at least one exemplary embodiment, step S104-1 may be implemented by:
Step S104-1-1, respectively extracting textures corresponding to the current frame image and the background frame image according to the edge gradients of the current frame image and the background frame image;
Step S104-1-2, determining that a video occlusion phenomenon occurs in a monitoring picture of the monitoring device under the condition that the texture corresponding to the current frame image is smaller than the texture corresponding to the background frame image.
The texture corresponding to the image is obtained through the edge gradient extraction mode, so that more accurate texture information can be obtained, and efficient and accurate video occlusion judgment is realized.
In order to prevent false detection due to subtle differences between images, a certain error threshold may be set, for example, in at least one exemplary embodiment, a case where the texture corresponding to the current frame image is smaller than the texture corresponding to the background frame image may include: the texture X c corresponding to the current frame image and the texture X b corresponding to the background frame image meet the condition X b-Xc≥θ1, wherein θ 1 is a texture difference judgment threshold value and the value is larger than 0.
In order to prevent false detection of occlusion caused by close-distance passing of a large object, an auxiliary scheme for judging repeated background may be provided, for example, in at least one exemplary embodiment, in a case that it is determined that a video occlusion phenomenon occurs on a monitoring screen of the monitoring device, the method may further include: acquiring data X' m of a frame of image shot before a video shielding phenomenon occurs on a monitoring picture of the monitoring equipment; sequentially determining a difference value eta=x ' m-X'j between data X ' j and data X ' m of the N frames of images after the video occlusion phenomenon occurs, wherein j=1, 2, … N, and N is a preset number; and filtering out a determination result of the video shielding phenomenon of a monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that a difference value eta corresponding to the picture exists in the N frames of pictures and is smaller than a preset difference value threshold. That is, it is determined by this method that the large object passes closely, which results in the judgment that the monitoring device is blocked, because the event is transient, the blocking phenomenon does not need to be reflected in the screen abnormality detection result.
In at least one exemplary embodiment, step S104-2 may be implemented by:
Step S104-2-1, extracting a first characteristic point in the current frame image and a second characteristic point in the background frame image;
Step S104-2-2, calculating a first feature descriptor of each first feature point in a preset area corresponding to each first feature point in the current frame image, and calculating a second feature descriptor of each second feature point in a preset area corresponding to each second feature point in the background frame image;
Step S104-2-3, determining pairing feature points between the current frame image and the background frame image according to the first feature descriptors and the second feature descriptors;
step S104-2-4, determining that a scene change phenomenon occurs in a monitoring picture of the monitoring device under the condition that the distance between the paired feature points between the current frame image and the background frame image is determined to be larger than a preset distance threshold value.
According to the scheme, the pairing feature points can be rapidly determined through the feature descriptors corresponding to the feature points of the current frame image and the background frame image, and whether the scene change phenomenon occurs in the monitoring picture of the monitoring equipment is determined through determining whether the distance between the pairing feature points is larger than the preset distance threshold value or not.
In at least one exemplary embodiment, determining that the distance of the pairing feature point between the current frame image and the background frame image is greater than a predetermined distance threshold may include: determination ofWherein P j,c and P j,b are the positions of paired feature points between the current frame image and the background frame image, respectively, and K is the total logarithm of the paired feature points. The method can judge whether the sum of the distances of the paired feature points between the current frame image and the background frame image is larger than the preset distance threshold value or not in general, simplifies the calculation process of scene change, and can judge whether scene change occurs or not efficiently and accurately.
To prevent the problem of false alarms due to scene changes over a large number of foreground passes, such false alarms may be eliminated by updating the background in time, e.g. in at least one exemplary embodiment, in case it is determined that a scene change phenomenon occurs in the monitoring screen of the monitoring device, further comprising: acquiring a current frame image with the definition higher than a preset definition threshold (mainly to ensure the quality of a background image and enable a detection result to be more accurate), taking the acquired current frame image as a background frame image, and returning to execute the operation of determining whether the scene change phenomenon occurs in a monitoring picture of the monitoring equipment according to the position distance of the pairing characteristic points between the current frame image and the background frame image. By the scheme, the background can be updated in time, so that the judgment of scene change is more accurate, and false alarm caused by the scene change due to the fact that a large number of foreground passes is eliminated.
In order to prevent false detection of scene change caused by water surface or leaf shake, an auxiliary scheme for judging similarity between foreground and background may be provided, in at least one exemplary embodiment, in a case of determining that a scene change phenomenon occurs in a monitoring picture of the monitoring device, the method further includes: respectively expanding textures corresponding to the current frame image and the background frame image based on an expansion edge algorithm; determining the expansion result of the texture corresponding to the background frame imageExpansion result/>, of texture corresponding to the current frame imageDifference between/>And filtering out a determination result of the scene change phenomenon of the monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that the difference value rho is smaller than a preset expansion threshold value. Through the scheme, false detection of scene change caused by water surface or leaf shaking can be prevented, and the accuracy of judging the scene change is improved.
In order to increase the stability of detection and prevent the detection of an abnormality caused by an accidental situation, the abnormality determination may be performed only when a plurality of frames continuously generate an abnormality, for example, in at least one exemplary embodiment, when it is determined that the monitoring screen of the monitoring device generates the video occlusion phenomenon or the scene change phenomenon, the method further includes: determining whether the number of consecutive frame images in which the video occlusion phenomenon or the scene change phenomenon is detected to occur exceeds a predetermined frame number threshold; and filtering out the determination result of the video shielding phenomenon or the scene change phenomenon of the monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that the number of continuous frame images does not exceed the preset frame number threshold value. By the scheme, false detection caused by accidental conditions can be prevented, and the detection stability is improved.
From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of one or more embodiments of the present application may be embodied essentially or in part in the form of a software product stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and including instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present application.
Example 2
In this embodiment, a device for detecting an abnormality of a monitoring screen is further provided, and the device is used to implement the foregoing embodiments and preferred embodiments, and is not described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
Fig. 2 is a block diagram showing a configuration of a detection apparatus for monitoring screen abnormality according to embodiment 2 of the present application, as shown in fig. 2, the apparatus including:
An acquisition module 22 configured to acquire a current frame image captured by a monitoring device and a background frame image for performing anomaly detection on a monitoring screen of the monitoring device;
The anomaly detection module 24 is configured to perform anomaly detection on a monitoring picture of the monitoring device according to the texture difference and/or the position distance of the pairing feature points between the current frame image and the background frame image, so as to obtain a picture anomaly detection result.
By means of the device, abnormal conditions of the monitoring equipment can be rapidly identified based on the texture difference and the position distance of the paired characteristic points because the abnormal detection is carried out on the monitoring picture of the monitoring equipment according to the texture difference and/or the position distance of the paired characteristic points between the current frame image and the background frame image, and therefore the problem of how to timely find out abnormal phenomena (such as shielding of artificial or unexpected noise or scene change phenomenon) of video shielding, scene change and the like of the monitoring picture can be solved, and timely and rapid detection of the monitoring picture abnormality is achieved.
Fig. 3 is a detailed block diagram of the structure of the monitoring screen abnormality detection apparatus according to embodiment 2 of the present application, and as shown in fig. 3, the abnormality detection module 24 of the apparatus includes:
A video occlusion detection sub-module 242 configured to determine whether a video occlusion phenomenon occurs in a monitoring screen of the monitoring device according to the texture difference between the current frame image and the background frame image;
The scene change detection sub-module 244 is configured to determine whether a scene change phenomenon occurs in a monitoring screen of the monitoring device according to the position distance of the pairing feature point between the current frame image and the background frame image.
The apparatus of this embodiment is used to implement the method of embodiment 1, and the specific implementation process is described in embodiment 1, which is not repeated in this embodiment.
It should be noted that each of the above modules may be implemented by software or hardware, and for the latter, it may be implemented by, but not limited to: the modules are all located in the same processor; or the above modules may be located in different processors in any combination.
Example 3
An embodiment of the application also provides a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.
Alternatively, in the present embodiment, the above-described storage medium may be configured to store a computer program for performing the steps of:
step S1, acquiring a current frame image shot by monitoring equipment and a background frame image used for carrying out anomaly detection on a monitoring picture of the monitoring equipment;
And S2, performing anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference and/or the position distance of the pairing feature points between the current frame image and the background frame image to obtain a picture anomaly detection result. Optionally, the storage medium is further arranged to store a computer program for performing the steps of:
Alternatively, in the present embodiment, the storage medium may include, but is not limited to: a usb disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory RAM), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing a computer program.
Example 4
The embodiment describes in detail a specific processing procedure of the detection method of the monitoring screen abnormality. The scheme of the embodiment detects the picture of the monitoring equipment through the video shielding and scene change detection technology, judges whether shielding and scene change phenomena occur in the video so as to repair the equipment in time, and can adapt to object shielding of different degrees, different sizes and different colors to effectively distinguish scene change and video shielding.
In this embodiment, the scene change and the video occlusion are mainly determined by the similarity between the current frame and the background frame, so the algorithm involves the maintenance of the background frame and the calculation of the similarity. In order to process characteristic point changes caused by water surface shaking and partial shielding, calculating and processing differences of canny expansion of a background frame and a current frame; in order to deal with the false detection phenomenon caused before a large object (e.g., a large vehicle) passes through the monitoring camera, false detection can be reduced by sliding window filtering of N frames. And uses sharpness as an indicator of video occlusion, scene changes, background updates.
Fig. 4 is a detailed flowchart of a detection method of monitoring screen abnormality according to embodiment 4 of the present application, and a specific algorithm of abnormality detection is described in detail below in conjunction with fig. 4.
S401, judging textures of a current frame and a background frame, and firstly calculating canny edge gradients of the current frame and the background frame:
Xi＝Φ(xi) i＝c,b
wherein X is an extracted texture map, phi (-) is a canny edge extraction algorithm, X is an original gray level image, and c and b respectively represent a current frame gray level image and a background frame gray level image; then defining a video occlusion judgment formula as follows:
Xb-Xc＜θ1
If the texture of the current frame map is smaller than the background image, a video occlusion phenomenon may occur.
And S402, extracting feature points and calculating similarity. When the phenomena of video shielding and scene change occur, the picture of the video is changed greatly, namely, the characteristic points of the current image and the background image are extracted for measurement.
Firstly, extracting a harris characteristic point for a current frame image and a background image:
Wherein f (x, y) is a harris feature point, I (x k,yk) is an x k、yk region of the image I, W is a sliding window size, and taylor expansion is performed on I (x+Δx, y+Δy):
then, the harris feature point is:
Wherein, Is the structure tensor. Constructing a harris response function by using the structure tensor M:
R＝det(M)-k(trace(M))2
where det (M) is a determinant of M, trace (M) is a trace of M, and k is a constant parameter. When R → +. In the case of infinity, the air conditioner is controlled, is an angular point; when R is- & gt-infinity, the edge is the edge; when |R|→0, the region is a smooth region.
Then, feature points of the current image and the background image can be calculated through the above formula, and a harr feature descriptor is calculated in a region around each feature point:
xi,k,yi,k∈Wi
Wherein F i is the ith feature point, For the harr feature calculation, W i is the window area around the i-th feature point.
And finally, calculating the paired characteristic points of the current frame and the background frame by using the characteristic descriptors of the characteristic points of the current frame image and the background frame image, and calculating the similarity by the distance between the paired characteristic points:
S is the similarity of the current frame and the background frame, and the smaller the similarity is, the more similar the description is; p j,c、Pj,b is the position of the matched current and background frame feature points calculated by F i, respectively.
S403, calculating the similarity of the foreground and the background by using the canny expansion map.
The canny graph is first expanded:
Wherein X i is a canny texture map, gamma (. Cndot.) is a canny expansion edge algorithm, C and b represent the gray level image of the current frame and the gray level image of the background frame respectively for the expanded canny texture map.
Then using the canny expansion results of the current frame and the background frame to carry out occlusion and scene change filtering:
if the rho value is smaller, the explanation is a scene such as water surface or leaf shake.
And S404, updating the background image. When scene change occurs and the definition is clear, background update is performed so as to solve the problem of false alarm of scene change when a large number of foreground passes.
Step S405, video occlusion and scene change post-processing. The method for processing the problem of video shielding and scene change false alarm of the large object passing through the monitoring equipment in a short distance comprises the following steps:
η＝X'm-X'j j＝1,2,…N
Wherein X m is the image data of the previous frame before the scene change occurs; x j is the j-th frame image data after the scene change occurs; if the value of η is very small in the N frames, this indicates that a duplicate background is present, possibly a large object passing closely through the monitoring device. Further, in order to further increase the stability of detection, smoothing processing for abnormality determination is performed using a plurality of frames that appear successively in post-processing.
Step S406, according to the results of steps S401, S402, S403 and S405, whether the frame in the monitoring picture has video occlusion or scene change can be obtained.
In summary, the scheme provides a completed detection flow of video shielding and scene change, and the internal maintenance background image is used for detecting the video shielding and the scene change, the feature points and the feature descriptors are used for detecting the video shielding and the scene change, the sliding window is used for filtering false detection of larger objects through monitoring equipment, and the texture features are used for detecting object shielding in different proportions. The scheme has the following advantages:
1. The shielding area is not limited, and the shielding object with any proportion can be detected.
2. The shielding object is not limited by the color of the shielding object, and the shielding object with any color can be detected.
3. Effectively distinguish video shielding and scene change phenomenon, reduced the false alarm rate.
4. The algorithm of the scheme has stronger robustness and can adapt to more prospects passing through monitoring equipment and larger objects passing through the monitoring equipment.
It will be apparent to one skilled in the art that the modules or steps of one or more embodiments of the application described above may be implemented in a general purpose computing device, they may be centralized in a single computing device, or distributed over a network of computing devices, or they may alternatively be implemented in program code executable by computing devices, such that they are stored in a memory device and, in some cases, executed in a different order than that shown or described, or they may be implemented as individual integrated circuit modules, or as multiple modules or steps within a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.
The above description is only of the preferred embodiments of the present application and is not intended to limit the present application, but various modifications and variations can be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the principle of the present application should be included in the protection scope of the present application.
Claims (12)
1. A method for detecting an abnormality in a monitor screen, comprising:
Acquiring a current frame image shot by monitoring equipment and a background frame image for carrying out anomaly detection on a monitoring picture of the monitoring equipment;
Performing anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points to obtain a picture anomaly detection result;
The abnormal detection of the monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing characteristic points comprises at least one of the following steps: determining whether a video shielding phenomenon occurs to a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image; and determining whether a scene change phenomenon occurs in a monitoring picture of the monitoring equipment according to the position distance of the pairing feature points between the current frame image and the background frame image.
2. The method of claim 1, wherein determining whether a video occlusion phenomenon occurs in a monitoring picture of the monitoring device based on the texture difference between the current frame image and the background frame image comprises:
extracting textures corresponding to the current frame image and textures corresponding to the background frame image according to the edge gradients of the current frame image and the background frame image respectively;
And under the condition that the texture corresponding to the current frame image is smaller than the texture corresponding to the background frame image, determining that a video shielding phenomenon occurs in a monitoring picture of the monitoring equipment.
3. The method of claim 2, wherein the texture corresponding to the current frame image is smaller than the texture corresponding to the background frame image comprises:
The texture X c corresponding to the current frame image and the texture X b corresponding to the background frame image meet the condition X b-Xc≥θ1, wherein θ 1 is a texture difference judgment threshold value and the value is larger than 0.
4. A method according to any one of claims 1-3, characterized in that in case it is determined that a video occlusion phenomenon occurs in a monitoring picture of the monitoring device, it further comprises:
Acquiring data X' m of a frame of image shot before a video shielding phenomenon occurs on a monitoring picture of the monitoring equipment;
Sequentially determining a difference value eta=x ' m-X'j between data X ' j and data X ' m of the N frames of images after the video occlusion phenomenon occurs, wherein j=1, 2, … N, and N is a preset number;
and filtering out a determination result of the video shielding phenomenon of a monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that a difference value eta corresponding to the picture exists in the N frames of pictures and is smaller than a preset difference value threshold.
5. The method according to claim 1, wherein determining whether a scene change phenomenon occurs in a monitoring screen of the monitoring device according to the position distance of the pairing feature point between the current frame image and the background frame image comprises:
extracting a first characteristic point in the current frame image and a second characteristic point in the background frame image;
Calculating a first feature descriptor of each first feature point in a preset area corresponding to each first feature point in the current frame image, and calculating a second feature descriptor of each second feature point in a preset area corresponding to each second feature point in the background frame image;
Determining pairing feature points between the current frame image and the background frame image according to the first feature descriptors and the second feature descriptors;
And under the condition that the distance between the pairing feature points between the current frame image and the background frame image is larger than a preset distance threshold value, determining that a scene change phenomenon occurs in a monitoring picture of the monitoring equipment.
6. The method of claim 5, wherein determining that a distance of a pairing feature point between the current frame image and the background frame image is greater than a predetermined distance threshold comprises:
determination of Wherein P j,c and P j,b are the positions of paired feature points between the current frame image and the background frame image, respectively, and K is the total logarithm of the paired feature points.
7. The method according to claim 1, wherein in case it is determined that a scene change phenomenon occurs in a monitoring screen of the monitoring device, further comprising:
And acquiring a current frame image with the definition higher than a preset definition threshold, taking the acquired current frame image as a background frame image, and returning to execute the operation of determining whether the scene change phenomenon occurs on a monitoring picture of the monitoring equipment according to the position distance of the pairing characteristic points between the current frame image and the background frame image.
8. The method according to any one of claims 1, 5-7, further comprising, in the event of determining that a scene change phenomenon occurs on a monitoring screen of the monitoring device:
respectively expanding textures corresponding to the current frame image and the background frame image based on an expansion edge algorithm;
determining the expansion result of the texture corresponding to the background frame image Expansion result/>, of texture corresponding to the current frame imageDifference between/>
And filtering out a determination result of the scene change phenomenon of the monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that the difference value rho is smaller than a preset expansion threshold value.
9. The method according to any one of claims 1-8, further comprising, in case it is determined that the video occlusion phenomenon or the scene change phenomenon occurs on a monitoring screen of the monitoring device:
determining whether the number of consecutive frame images in which the video occlusion phenomenon or the scene change phenomenon is detected to occur exceeds a predetermined frame number threshold;
And filtering out the determination result of the video shielding phenomenon or the scene change phenomenon of the monitoring picture of the monitoring equipment from the picture abnormality detection result under the condition that the number of continuous frame images does not exceed the preset frame number threshold value.
10. A monitoring screen abnormality detection apparatus, comprising:
the acquisition module is used for acquiring a current frame image shot by the monitoring equipment and a background frame image used for carrying out anomaly detection on a monitoring picture of the monitoring equipment;
The anomaly detection module is used for carrying out anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points to obtain a picture anomaly detection result;
the abnormality detection module includes: the video shielding detection sub-module is used for determining whether a video shielding phenomenon occurs on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image; and the scene change detection sub-module is used for determining whether a scene change phenomenon occurs in a monitoring picture of the monitoring equipment according to the position distance of the pairing characteristic points between the current frame image and the background frame image.
11. A storage medium having a computer program stored therein, wherein the computer program is arranged to perform the method of any of claims 1 to 9 when run.
12. A detection system for monitoring screen anomalies, comprising:
The one or more monitoring devices are arranged to collect monitoring video streams and transmit the monitoring video streams to the video monitoring server;
The video monitoring server is connected with the one or more monitoring devices and comprises a memory and a processor, wherein the memory stores a computer program, and the processor is used for running the computer program to execute the following operations:
receiving the monitoring video stream transmitted by the monitoring equipment;
acquiring a current frame image in the monitoring video stream and a background frame image for carrying out anomaly detection on a monitoring picture of the monitoring equipment;
Performing anomaly detection on a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing feature points to obtain a picture anomaly detection result;
The abnormal detection of the monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image and/or the position distance of the pairing characteristic points comprises at least one of the following steps: determining whether a video shielding phenomenon occurs to a monitoring picture of the monitoring equipment according to the texture difference between the current frame image and the background frame image; and determining whether a scene change phenomenon occurs in a monitoring picture of the monitoring equipment according to the position distance of the pairing feature points between the current frame image and the background frame image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010676575.0A CN111898486B (en) | 2020-07-14 | 2020-07-14 | Monitoring picture abnormality detection method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010676575.0A CN111898486B (en) | 2020-07-14 | 2020-07-14 | Monitoring picture abnormality detection method, device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111898486A CN111898486A (en) | 2020-11-06 |
| CN111898486B true CN111898486B (en) | 2024-05-10 |
Family
ID=73192712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010676575.0A Active CN111898486B (en) | 2020-07-14 | 2020-07-14 | Monitoring picture abnormality detection method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111898486B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112637568B (en) * | 2020-12-24 | 2021-11-23 | 中标慧安信息技术股份有限公司 | Distributed security monitoring method and system based on multi-node edge computing equipment |
| CN112597952A (en) * | 2020-12-28 | 2021-04-02 | 深圳市捷顺科技实业股份有限公司 | Method, device and system for identifying monitoring state of camera and storage medium |
| CN113205489A (en) * | 2021-04-16 | 2021-08-03 | 广东海文通信有限责任公司 | Monitoring image detection method and device, computer equipment and storage medium |
| CN114245200B (en) * | 2021-11-02 | 2024-05-07 | 浙江大华技术股份有限公司 | Monitoring scene change determining method, device, system and electronic device |
| CN113936242B (en) * | 2021-12-14 | 2022-03-11 | 苏州浪潮智能科技有限公司 | Video image interference detection method, system, device and medium |
| CN116563429B (en) * | 2023-06-14 | 2023-10-27 | 厦门华厦学院 | Animation design processing supervision system and method based on cloud computing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799863A (en) * | 2012-07-02 | 2012-11-28 | 中国计量学院 | Method for detecting group crowd abnormal behaviors in video monitoring |
| CN109040709A (en) * | 2018-09-21 | 2018-12-18 | 深圳市九洲电器有限公司 | Video monitoring method and device, monitoring server and video monitoring system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101932535B1 (en) * | 2012-08-27 | 2018-12-27 | 한화테크윈 주식회사 | Indoor surveillance system and method |
-
2020
- 2020-07-14 CN CN202010676575.0A patent/CN111898486B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799863A (en) * | 2012-07-02 | 2012-11-28 | 中国计量学院 | Method for detecting group crowd abnormal behaviors in video monitoring |
| CN109040709A (en) * | 2018-09-21 | 2018-12-18 | 深圳市九洲电器有限公司 | Video monitoring method and device, monitoring server and video monitoring system |
Non-Patent Citations (1)
| Title |
|---|
| 夏裕建 ; 孙涵 ; .监控视频异常检测与质量诊断.计算机应用与软件.2016,(第06期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111898486A (en) | 2020-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111898486B (en) | Monitoring picture abnormality detection method, device and storage medium | |
| US10417503B2 (en) | Image processing apparatus and image processing method | |
| CN110650316A (en) | Intelligent patrol and early warning processing method and device, electronic equipment and storage medium | |
| CN108197604A (en) | Fast face positioning and tracing method based on embedded device | |
| CN101167086A (en) | Human detection and tracking for security applications | |
| CN104966304A (en) | Kalman filtering and nonparametric background model-based multi-target detection tracking method | |
| US20110280478A1 (en) | Object monitoring system and method | |
| CN112102409A (en) | Target detection method, device, equipment and storage medium | |
| CN109255802B (en) | Pedestrian tracking method, device, computer equipment and storage medium | |
| CN112101195B (en) | Crowd density estimation method, crowd density estimation device, computer equipment and storage medium | |
| CN110414571A (en) | A kind of website based on Fusion Features reports an error screenshot classification method | |
| CN109030499B (en) | Device and method suitable for continuous online detection of target defects and preventing repeated counting of defect number | |
| CN113947754A (en) | Vision-based ship machinery running state monitoring method and system and storage medium | |
| DE112020005223T5 (en) | Object tracking device and object tracking method | |
| CN114674826A (en) | Visual detection method and detection system based on cloth | |
| CN110909617A (en) | Living body face detection method and device based on binocular vision | |
| CN114092515A (en) | Target tracking detection method, device, equipment and medium for obstacle blocking | |
| CN111563492B (en) | Fall detection method, fall detection device and storage device | |
| Wang et al. | Coarse-to-fine grained image splicing localization method based on noise level inconsistency | |
| CN112232113B (en) | Person identification method, person identification device, storage medium, and electronic apparatus | |
| CN112686162B (en) | Method, device, equipment and storage medium for detecting clean state of warehouse environment | |
| CN111985331B (en) | Detection method and device for preventing trade secret from being stolen | |
| CN114708544A (en) | Intelligent violation monitoring helmet based on edge calculation and monitoring method thereof | |
| Miljanovic et al. | Detection of windows in facades using image processing algorithms | |
| CN113111777A (en) | Indoor personnel density detection system and detection method based on ARM platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |